Method for preventing outflow of idea originality in automatic program generation function

Abstract

[PROBLEM] When generative AI or automatic programming which does not require human coding become mainstream, the code which automatic program generation functions create using similar natural language inputs will become similar and it will be easy for others to imitate the code. The purpose of the present invention is to realize automatic programming that ensures originality. [SOLUTION] Code or the like, which is paired with a word, a clause, a sentence, grammar, etc. and is stored in a database or the like associated with an automatic program generation function, is divided into a shared part and a user-specific part, thereby controlling use by another person.

Description

How to prevent original ideas from leaking out in automatic program generation functions

[0001] The present invention relates to an automatic program generation function that automatically creates software, so-called automatic programming, and to an environment that executes multiple automatic programs, i.e., a technology that is effective when applied to the automatic program generation function.

[0002] As generative AI becomes more widely used and automatic program generation functions that do not require human programming—known as automatic programming—become mainstream, it is believed that programs will be generated by artificial intelligence using natural language processing, reducing the need for human programming. Currently, automatic programming generates programs using associations between natural language and code created by humans. Programming code for natural language is associated within a database associated with the automatic program generation function or within a program with the automatic program generation function. Corresponding code is added for new words and sentences. This code is inserted within the program with the automatic program generation function or within a commonly used database. This results in common learning and growth, regardless of the natural language input by each user, and all users can utilize the relationship between the learned code and natural language. This structure, which has natural language-compatible code within a database, allows multiple databases to be sequentially synchronized and use the same database. Furthermore, a structure with natural language-compatible code within an automatic program generation function allows the same automatic program generator to be maintained by adding learned results to all automatic program generation functions, recompiling them, and sequentially synchronizing them. This allows all users to enjoy the same learning results.

[0003] On the other hand, a disadvantage of this uniform environment for automatic program generation is that using the same natural language scenario can result in the generation of identical programs. Even if a unique idea is realized and an application software with added value is developed, similar application software can be developed by providing similar natural language sentences, resulting in a loss of competitiveness. Furthermore, when automatic program generation functions, such as generative AI, have learning capabilities, a program created from a natural language sentence provided to one automatic program generation function can be shared by all automatic program generation functions, potentially resulting in the loss or leakage of the originality of the idea.

[0004] JP 09-073306: A method for inputting a database closely related to program input, by making the input method for creating a program closer to natural language (several types proposed) JP 06-214776: Automatically generating the target program source by inputting software specifications written in natural language JP 7064680: Adding a natural language processing method to JP 06-214776, storing the corresponding program code in a database JP 6958884: Similar in content to JP 06-21477, but adding a conversation reception function

[0005] In systems and programs with automatic program generation functions, including AI, when the same automatic program generation function is used, inputting any similar natural language sentence in different usage environments will result in the generation of similar program code. This means that the unique ideas in the input natural language sentence will leak out through the similar program code that is generated. To avoid this, systems with automatic program generation functions and program generation functions themselves are designed to automatically generate both common program code and unique program code that cannot be imitated by others.

[0006] Here, the program code paired with natural language refers to the source code or compiled program of the program paired with words, phrases, sentences, grammar, etc. stored in the automatic program generation function itself or in an associated database or file system, and refers to the extraction of program code paired with words, phrases, sentences, and grammar divisions of an input sentence using natural language processing from a database or file system associated with the automatic program generation function or from those stored within the automatic program generation function.

[0007] The automatic program generation function has program code corresponding to words, phrases, sentences, grammar, etc. stored in a database or file system commonly associated with the automatic program generation function, a compiled program, or program code corresponding to words, phrases, sentences, grammar, etc. incorporated into the automatic program generation function program itself, and a compiled program. Separately, for each user or user group, there is program code corresponding to words, phrases, sentences, grammar, etc. stored in a database or file system associated with a separate automatic program generation function, a compiled program, or program code corresponding to words, phrases, sentences, grammar, etc. incorporated into the automatic program generation function program itself, and a compiled program, which makes it impossible for others to use the program code corresponding to highly unique or confidential words, phrases, sentences, grammar, etc. and the compiled program.

[0008] A general AI flow example in which the content given by the natural language processing of so-called automatic programming is divided and the code corresponding to the content is extracted from the database, including a database or file system from the automatic program achievement function, or a database or file system or a pair of natural language and program code that is generated by the automatic program achievement function, which controls access for each user separately from the common part of the pair of natural language and program code that is used in the automatic program achievement function, and an example of code extraction when generated by AI. A condition is sent from one automatic program generation function to an unspecified number of or arbitrary multiple automatic program generation functions, and codes that match the condition are extracted. An example diagram of an automatic program generation function that asks whether it has the code or data that meets the conditions, and responds if it has code or data that meets the conditions.If the automatic program generation function, or so-called automatic programming, does not have the content divided by natural language processing in its database, it checks with another automatic program generation function, or so-called automatic programming, and if it does, it asks them to send the content.An example diagram of a flow in which an automatic program generation function has subdata linked to the code or data it possesses.An example diagram of code exchange after one automatic program generation function in Figure 1 responds to given conditions.An example diagram of data exchange after one automatic program generation function in Figure 1 responds to given conditions.

[0009] Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. However, the present invention can be implemented in many different forms and should not be interpreted as being limited to the description of the present embodiment. Note that the same elements will be assigned the same numbers throughout the embodiments.

[0010] In the following embodiments, a method or a system will be mainly described, but as will be apparent to those skilled in the art, the present invention can also be implemented as a program usable by a computer. Therefore, the present invention can take the form of hardware, software, or a combination of software and hardware. The program can be recorded on any computer-readable medium, such as a hard disk, a DVD-ROM, an optical storage device, a magnetic storage device, or a semiconductor storage device.

[0011] In addition, the following embodiments may use general computer systems, etc. Computer systems, network systems, and systems for artificial intelligence that can be used in the embodiments include hardware resources typically found in computer systems, network systems, and systems for artificial intelligence, such as a central processing unit (CPU), a main memory (RAM), a non-volatile memory (ROM), a coprocessor, a graphics accelerator (GPU), cache memory, an input / output control device (I / O), semiconductors for artificial intelligence functions, and flash memory. They may also include external storage devices such as hard disk drives and flash memory, and communication means connectable to networks such as the Internet. Computer systems include various types of computers, such as personal computers, workstations, mainframe computers, and servers, as well as cloud-based network-connected computers, tablet computers, various portable information terminals such as smartphones, and systems that constitute artificial intelligence.

[0012] The present invention aims to prevent the output of similar program code even when the input content is highly unique by adding program code generated by the automatic program generation function to the automatic program generation function, or so-called automatic programming, so that the learning and growing generation AI and automatic program generation function will be able to output similar programming code when similar natural language sentences are input as a result of growth, thereby preventing the violation of confidentiality and the leakage of uniqueness.

[0013] The configuration of the present invention, as shown in Figure 1, comprises a database, file system, or internal structure of the automatic program generation function that stores a combination of natural language and its corresponding program code that can be used by all users and is associated with the automatic program generation function, and a separate structure within the database, file system, or automatic program generation function that stores a combination of natural language and its corresponding program code that can be used only by authorized users by controlling access for each user. The natural language in the user common portion refers to each word, phrase, sentence, and grammar, either singly or in combination, and stores a combination of its corresponding program code. The natural language in the portion available only to authorized users is basically stored separately from the common portion, storing pairs of arbitrary natural language sentences entered by users or user groups who own the database, file system, or internal structure of the automatic program generation function, and program code composed of those sentences. Program code automatically programmed using arbitrary natural language sentences entered by users, i.e., code that reflects their own intentions, cannot be used by others without permission. Functions are realized by combining configurations of smaller functions. Therefore, if the use of small functions is not permitted to others, it becomes difficult to compose unique functions that are created by composing larger functions. Therefore, it is not possible to generate a program with the same structure unless the natural language documents that compose all functions are input exactly the same and the combination of natural language and code in the user-specific part is permitted. This makes it possible to prevent the leakage of unique ideas to others through the program.

[0014] When creating a code for input of a natural language sentence, phrase, grammar, etc., the user first checks whether the same natural language sentence, phrase, grammar, etc. already exists in the user's own user-specific database, file system, or automatic program generation function, and if so, uses the corresponding code. If not, the user sequentially checks for permission to use from all other users. If one or more other users have given permission to the user, the user checks whether the natural language sentence of the content description and the input natural language sentence exist in the other users' user-specific databases, file systems, or automatic program generation functions. If so, the user receives the stored combination of the content description and code and format, checks the stored content against the input natural language sentence, phrase, grammar, etc., and if they are identical, uses this code. If not present in any other users' databases, the user obtains and combines the basic code paired with the natural language from the user's common basic database, combining the basic form of the natural language and the basic code. The basic form of natural language refers to the input natural language that, if condition A, performs process B, otherwise performs process C, and the basic code refers to general automatic programming that assigns codes starting from a basic form such as if condition A process B else process C.

[0015] When automatic program generation functions, such as generative AI, process natural language text as shown in Figure 2, they typically divide the input natural language text by content, further subdivide it into sentences and phrases, and then extract and synthesize codes corresponding to the words and grammar stored in a database, file system, or the automatic program generation function program itself based on the relationships between words and grammar to create code. The generated codes are then stored in a commonly used database, file system, or automatic program generation function program itself as pairs with the decomposed sentences and subdivided sentences, sentences, and phrases, allowing the automatic program generation function to learn. As a result, when a similar input text is given, it becomes possible to extract codes corresponding to sentences, sentences, and phrases created by others, thereby allowing others to use the code they have created. To avoid this, the code is stored in a database, file system, or automatic program generation function program itself, which is a private area for the inputter, and is accessible only to authorized users. This allows access only to specific users, preventing the leakage of original ideas to others. Normally, when creating a program, a larger program is constructed by combining smaller unit programs. Therefore, unless the code that corresponds to the detailed functional content is constructed in the common part, there is no risk that the code that performs the function as a whole will be imitated by inputting the functional overview.

[0016] Figure 3 shows an example of the exchange of data and code for self-learning and self-development between multiple automatic program generation functions, including a generation AI. If an automatic program generation function does not have the code for the desired natural language sentence or phrase in its associated database or file system, or in the automatic program generation function itself, it checks whether the code corresponding to the natural language sentence or phrase is available in the non-common parts of other automatic program generation functions, i.e., in the user's own database, file system, or the automatic program generation function itself. If the automatic program generation function whose possession is confirmed has already granted access to the user who owns the automatic program generation function that asked the question, and determines that the natural language content of the question matches the description that pairs with the code it possesses, it responds to the automatic program generation function that made the question that it possesses the code, and sends the code, format, and content description. The person who asked the question determines whether the description matches their question, and if it does, incorporates it into their own database, file system, or automatic program generation function itself, thereby growing. It checks for any adverse effects that may be caused by incorporating or using the code.

[0017] FIG. 4 shows an example of a flow from when automatic program generation function A receives a natural language sentence to be realized in FIG. 3, to when it checks with automatic program generation functions B, C, and D and receives the corresponding program code.

[0018] FIG. 5 shows an example of an interface for the database or file system of the automatic program generation function or the part that stores code within the automatic program generation function. In addition to codes for input natural language sentences, phrases, words, grammar, or combinations thereof, the example stores the input, output, or combined format of the code, and the stored sentences, phrases, words, grammar, or combinations thereof that explain the content. For example, when a user is asked to process a natural language sentence by another person, if the user is able to respond to a possession confirmation, the user can reply with the input, output, or combined format and a content description. If a match is determined, the user can return the code to the authorized user. Another example shows a case where a user possesses processed data other than the code. In response to a question about possession of arbitrary data, the user can return a content description, data structure, and data, allowing the authorized user to use the arbitrary data.

[0019] FIG. 6 shows an example of code exchange between two automatic program generation functions in the description of 0017, and similarly FIG. 7 shows an example of data exchange between two automatic program generation functions applied to the description of 0017.

[0020] When receiving code or compiled programs from other users, in order to prevent system errors that may occur after import due to errors or viruses in the code or compiled programs, by checking the code or compiled programs to be imported for errors at the time of import, it is possible to prevent errors after import from affecting the entire program after import.

[0021] It can be applied to generative AI, which allows each individual or company to build their own unique application software that prevents leakage to others.

[0022] In addition, the uniquely generated program code can be used on sites that sell it for a fee, and by applying it to a generation AI that provides code for created functions only to registered users, it becomes possible to give permission to use unique know-how only to specific users.

Claims

1. A computer system having a generation AI or automatic program generation function that has one or both of the combinations of code based on natural language sentences and natural language sentences, or the combinations of compiled programs based on natural language sentences and natural language sentences, which can be used in common with other users, and that has one or both of the combinations dedicated to the user's own use, and that can incorporate the code and compiled programs of other users whose existence has been confirmed by querying the automatic program generation function of other users into its own automatic program generation function.

2. A generating AI or computer system with an automatic program generation function that can grant the right to use the code or compiled program exclusive to the user in response to a request from another person, as described in claim 1.

3. A computer system having a generating AI or program automatic generation function with input / output formats or a combination of input / output formats and code content descriptions in association with the user's own dedicated code or compiled program as described in claim 1.

4. A computer system having a generating AI or program automatic generation function with a data structure or a combination of a data structure and a data content description for data related to the code or compiled program for the user's own exclusive use as described in claim 1.

5. A computer system having a generation AI or an automatic program generation function as described in claim 1, which has the function of inquiring from its own automatic program generation function to another automatic program generation function about the presence or absence of data related to conditions specified in the other automatic program generation function.

Images